Naoto Usuyama, Jeya Maria Jose Valanarasu, Sicong Yao, Hanwen Xu, Jaspreet Bagga, Guanghui Qin, Robert E. Kramer, Cliff Wong, Soohee Lee, Hao Qiu, Theodore Zhengde Zhao, Racheli Ben Shimol, Angela Crabtree, Kevin Matlock, Eduardo Alejandro Lozano Garcia, Naiteek Sangani, Alberto Santamaria-Pang, Maximilian Rokuss, Yashna Hasija, Naisargi Manishkumar Patel, Jason Entenmann, Alexandra Q. Bartlett, Bill J. Wright, Bernard A. Fox, Brian Piening, Sheng Zhang, Sheng Wang, Tristan Naumann, Carlo Bifulco, Hoifung Poon
机构
*
Microsoft Research(微软研究院)
;
Paul G. Allen School of Computer Science and Engineering, University of Washington(华盛顿大学保罗·G·艾伦计算机科学与工程学院)
;
Providence Genomics(普罗维登斯基因组学公司)
;
Earle A. Chiles Research Institute, Providence Cancer Institute(普罗维登斯癌症研究所厄尔·A·奇尔斯研究所)
;
Providence Research Network(普罗维登斯研究网络)
CommentsAccepted to ICML 2026. The title has been changed from "From Correction to Mastery: Reinforced Distillation of Large Language Model Agents" to "Student-Centered Distillation Narrows the Agentic Gap Between Small and Large LLMs"; the camera-ready version has been uploaded
Break Through the Compression Bottleneck: From Theory to Practice
突破压缩瓶颈:从理论到实践
Xiusheng Huang, Lu Wang, Yequan Wang, Jun Zhao, Kang Liu
机构
*
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所)
;
University of Chinese Academy of Sciences(中国科学院大学)
;
Beijing Academy of Artificial Intelligence(北京人工智能研究院)
专题命中
效率与部署
:large language model(abstract);language model(abstract);分类 cs.CL、cs.AI
Comments64pp, LaTeX. v2 rebuilds arXiv:2606.31796 into a full report: matched-compute discrimination/generation dissociation, recurrent-yield compression (48->6)+MoEE, pre-registered 1B->13.7B H-PRESERVE, sovereign Korean tokenizer (+9.2% vs Gemma-4), government HLE(Ko) column lead, cyber specialization. Recurrent compression cited by Loopie (arXiv:2607.16051). Tables incl.; v1 in history